American Economic Review: Papers & Proceedings 2011, 101:3, 233–237
http://www.aeaweb.org/articles.php?doi=10.1257/aer.101.3.233
Moving Beyond Deregulation †
Why Can’t US Airlines Make Money?
By Severin Borenstein*
The US domestic airline industry was effec-
tively deregulated in the fall of 1978 with elimi-
nation of most economic restrictions on new
entry and pricing. The industry lost $10 billion
from 1979 to 1989, made $5 billion in the 1990s
and lost $54 billion from 2000 to 2009 (all fig-
ures in 2009 dollars).1 To put these figures in
context, at the end of 2000, after six consecu-
tive profitable years, the entire book value of US
passenger carriers’ assets was $159 billion and
shareholder equity was $40 billion.
This dismal financial record isn’t what econo-
mists, analysts, or industry participants predicted
in 1978. It is a puzzle to industrial organization
economists and a challenge to the views of dereg-
ulation advocates. The puzzle is compounded by
the fact that the industry saw robust investment
until 2001 and has seen only modest disinvest-
ment in the financially disastrous 2000s. From
1979 to 2001, the US airline passenger fleet grew
in every year, by an average of 4.9 percent per
year measured by aircraft and 3.6 percent per
year measured by aircraft-seats. After peaking in
†
Discussants: Catherine Wolfram, University of California-
Berkeley and NBER; Nancy L. Rose, Massachusetts Institute
of Technology.
* Haas School of Business, University of California,
Berkeley, CA 94720-1900 (e-mail: borenste@haas.berke-
ley.edu). For helpful comments, I thank Silke Forbes,
Will Gillespie, Ken Heyer, Paul Joskow, Nancy Rose, and
Catherine Wolfram. This paper is dedicated to the memory
of Alfred E. Kahn, who passed away on December 27, 2010.
I was lucky enough to work for Fred at the Civil Aeronautics
Board in 1978 and to speak with him occasionally since
then about the airline industry and government regulation.
His approach to industrial organization and regulation, and
the application of research to nonpartisan policymaking,
set a standard to which all IO economists should aspire.
His insights continue to influence the best research on eco-
nomic regulation. He was the very model of a modern (and
thoughtful) IO economist.
1
I focus on net income before extraordinary charges and
gains. Including such adjustments doesn’t alter the basic
analysis but for some carriers causes large profit swings
from one year to the next. For a more detailed description
of the analysis reported here, see Severin Borenstein (2011).  after the September 11 attacks, the tax burden has
233
2001, aircraft and aircraft-seats declined from the
end of 2001 to the end of 2008, 1.7 percent and
1.4 percent per year respectively.
There is no conventional long-run equilib-
rium explanation for an industry that perpetu-
ally loses money, but there are a number of
disequilibrium theories that have been suggested
by industry participants, financial analysts, and
researchers. In this short paper I discuss some of
these theories and attempt to narrow down the
range of plausible explanations.
I.  Exogenous Cost Drivers: Taxes and Fuel
Industry leaders argue that taxes on airline
tickets have risen drastically and are a sig-
nificant contributor to the airlines’ losses. The
ticket tax today includes a 7.5 percent excise tax
and fees of $6.20 per segment flown. In addi-
tion, many airports impose passenger facilities
charges (PFCs) of up to $4.50 on each passenger
boarding a flight at the airport. One can argue
about whether these taxes are excessive given
the government costs of supporting the industry,
but it is difficult to see how these would lead to
long-run losses. The average tax (including fed-
eral ticket taxes and PFCs) as a percentage of
the base ticket price has climbed steadily and is
today about twice as high as when it was 8 per-
cent through most of the 1980s. But the average
dollar tax per ticket is today about $43 ($2009),
just a dollar or two more than it was in the late
1990s, the industry’s most profitable years.
Over the last 30 years, the form of taxation has
changed. In the 1980s, the entire ticket tax was a
percentage of the ticket value. Today, about half of
ticket tax revenues come from fixed per-­segment
fees. PFCs were added in the early 1990s, the
segment tax in 1997, and the September 11 secu-
rity fee in early 2002, all based on the number
of flights the passenger boards, regardless of the
fare paid. As a result, as real fares have declined
28 percent since 1992, dropping significantly
234 AEA PAPERS AND PROCEEDINGS
increased as a percentage of the base fare. But the
implied shift of demand (or marginal cost) that
airlines face due to ticket taxation is today about
the same as it was just before 9/11.
While taxes and fees have changed incremen-
tally, the industry scale has changed massively.
In the standard long-run adjustment dynamics,
it seems that the industry should have been able
to achieve the scale change necessary to incor-
porate and pass through these taxes. My own
research (in progress) suggests that changes in
passenger facilities charges are nearly entirely
passed through to customers within two quarters.
Fuel cost increases have certainly been a
significant component of losses in some years,
most obviously 2008. Over the deregulation
era, however, oil costs were highest in the first
seven years and the most recent five years—over
$40 per barrel in 2009 dollars—but in the 19
intervening years—1986 to 2004—real oil and
jet fuel prices were relatively stable and much
lower than in the early period of deregulation.
Yet, the industry still lost money in 13 of those
19 years and on net lost $31 billion ($2009).
When shocks do occur, there doesn’t appear to
be any barrier to capacity adjustment over three
to six months in response, as occurred in the
second half of 2008. Still, whether in response
to higher taxes or oil prices, reducing flight
schedules doesn’t eliminate costs if those costs
are fixed or sticky. In times of growing demand,
carriers can adjust fairly smoothly to unantici-
pated cost increases by growing more slowly,
without having to ground aircraft or reduce work
force size. When demand is stagnant or declin-
ing, however, rescaling operations in response to
upward cost shocks is more difficult and costly.
II.  Exogenous Demand Shocks
The role of demand shocks in airline losses
is most notable in 2001–02 and in 2008–09.
Prior to 9/11, however, it appears that domestic
demand grew fairly steadily. Inferring demand
shifts from average price (adjusted for trip dis-
tance) and revenue passenger-miles, demand
changes are presented in Figure 1 along with
the change in US real GDP for comparison.2
2
Following Borenstein and Nancy L. Rose (2007),
demand is assumed to be Q = AP ε, where Q is domestic
revenue passenger-miles, P is an index of domestic yield
MAY 2011
240%
Airline demand Real GDP
220%
200%
180%
160%
140%
120%
100%
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
Figure 1. Airline Demand and Real GDP,
Relative to 1979
Demand increased by 110 percent from 1979
to 2000, growing in 16 of those 21 years. Yet,
the industry made money in only eight of those
years and overall lost $3 billion (2009 dollars)
over this period. The intermittent economic
downturns during this period certainly affected
airline industry profits, but it’s very unlikely
that investors expected demand growth would
be completely constant and steady. It is hard to
see how unanticipated demand shocks during
this time could be a credible explanation for the
overall poor performance before 9/11.
Demand shocks are a more plausible explana-
tion for the losses of the 2000s. The post-9/11
demand drop, which was about 20 percent from
2000 to 2002, was unprecedented. By 2008,
demand was still about 3 percent lower than it had
been in 2000, and then it fell 11 percent in 2009.
Because of the fixed capital costs and sticky
labor costs, the decade of depressed demand was
accompanied by a decade of depressed prices. In
real terms, prices were 20 percent lower in 2009
than in 2000 despite the fact that jet fuel prices
were about $0.59 per gallon (52 percent) higher,
which, based on 2009 revenue passenger-miles
per gallon of fuel, raised overall costs by about
9 percent.
III.  Entry and Expansion of Low-Cost Carriers
Many industry observers and participants
point to low-cost (and low-fare) carriers (LCCs)
(adjusted for trip distance) and ε = −1. Figure 1 tracks A
over time. See Borenstein (2011) for details. 
VOL. 101 NO. 3
700,000
600,000
500,000
400,000
300,000
200,000
100,000
0
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
Legacy & regional carriers
Figure 2. Aircraft-Seats in Fleet
as part of the reason for low industry profits,
but there is wide disagreement on what the con-
nection is. If LCCs are simply offering a lower-
quality product, then their differentiated product
should find its niche in the market if there is
sufficient demand for that quality level, yielding
an equilibrium with both types earning normal
returns.
Among industry and labor leaders, a com-
mon view is that new low-cost entrants and
LCC incumbents have made excessive capac-
ity investments during growth periods, and
sometimes even during downturns, that have
depressed prices for all. In order to discourage
excessive investment, the largest airline pilots
union has called for increasing capital require-
ments as part of FAA licensing of new airlines.
But the evidence doesn’t appear to support the
idea that new entrants or older LCCs are more
prone to overinvestment than the legacy airlines.
Figure 2 presents the aircraft-seat fleet size of
LCCs and non-LCCs (including legacy carri-
ers and regional carriers who generally operate
as codeshare partners to the legacy carriers). It
shows that LCCs in aggregate have experienced
no more erratic fleet size adjustments despite
being less well-established on average.3 In fact,
they continued to grow gradually even after 9/11
while remaining much less unprofitable than the
legacy carriers, as shown below. If anything, it
appears to be the legacy carriers who are more
3
The declines in 1987, 1988, and 2007 were due to a
legacy carrier absorbing an LCC.  impact of FFPs.
Why Can’t US Airlines Make Money? 235
prone to overinvestment relative to the growth of
their traffic. Figure 2 also shows that the changes
in LCC fleet size are dwarfed by the variation of
the non-LCC fleet, suggesting that LCC invest-
ment decisions have not been the primary driver
in industry capacity changes.
An alternate view of LCCs is that they have
been gradually chipping away at the entrenched
positions of legacy carriers that have much higher
costs. The change has been gradual, because the
legacy carriers are also protected by network
1999
2001
2003
2005
2007
marketing programs and other activities that
Low-cost carriers
raise barriers to entry by more efficient firms.
Potentially exclusionary activities of legacy car-
riers include frequent-flyer programs (FFPs)
and corporate discount programs that exchange
discounts for customer loyalty on a portfolio of
unrelated routes,4 as well as relationships with
airports that allow large incumbents to restrict
the availability of gates, landing slots, and other
resources to potential entrants.
LCCs have been growing steadily since
the early 1990s, from about 10 percent mar-
ket share in 1994 to about 24 percent in 2009,
with Southwest accounting for about half of
LCC traffic in most years. LCCs now compete
(defined as at least 10 percent passenger share)
on over 60 percent of all airport pairs. And LCCs
have maintained much lower costs than the leg-
acy carriers. Figure 3 shows that, adjusted for
the average flight distance, legacy carrier costs
have remained 30 percent to 60 percent higher
than the LCCs’ for nearly all of the deregulation
era, averaging about 40 percent higher in the last
decade.
While the cost differential between LCCs
and non-LCCs has remained large, the average
price differential has been shrinking, as shown
in Figure 4. Figure 4 is adjusted for the average
trip distance of passengers flying on each type of
carrier. LCC fares have declined much less than
those of legacy carriers in the 2000s, reflecting
their lower burden of excess aircraft capacity.
This is no doubt a large part of the reason that
LCCs have suffered much milder losses in the
2000s, as shown in Figure 5.
4
Borenstein (1996) discusses the potential anticompeti-
tive effects of such repeat-buyer programs in more detail.
Mara Lederman (2007, 2008) presents evidence on the
 
236 AEA PAPERS AND PROCEEDINGS MAY 2011

20 1.5
18
1
16
14 0.5
12
0
10

2003

2005

2007

2009
1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001
8 −0.5
6
−1
4
2 −1.5
0
−2
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
−2.5
Legacy & regional carriers Low-cost carriers Legacy & regional carriers Low-cost carriers

Figure 3. Operating Cost per Available Seat-Mile Figure 5. Net Income per Available Seat-Mile
(2009 cents, adjusted for flight distance) (2009 cents)

last 20 years.5 As a result, while the exogenous

demand and cost shocks have affected all carri-
100% ers, the legacy airlines have fared much worse
90% financially, and LCCs have grown steadily.
80% The response of legacy carriers has been to
70% expand their networks through mergers and alli-
60% ances. There is little evidence that such moves
50% narrow the cost gap with LCCs, but network
40%
expansion may help differentiate their prod-
30%
ucts and improve service. It also may increase
their ability to use network marketing devices to
20%
dampen LCC competition.6
10%
The financial results for legacy airlines and
0% LCCs have improved substantially in 2010, and
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009

the industry seems likely to be closer to break-

even on domestic operations. Still, the experi-
Figure 4. Price Premium of Legacy Carriers ence of the last decade suggests that until legacy
(adjusted for trip distance)
carriers can either close the cost gap with LCCs
or increase the price premium they can maintain,
they will likely have difficulty earning consis-
IV.  A Series of Unfortunate Events? tent profits through the typical cycles in the air-
line business environment.
Demand and cost shocks have certainly This short paper obviously doesn’t settle the
played a significant role in the airline industry’s issues surrounding airline profitability. The topic
poor financial results, but there is little reason would benefit from much more investigation by
to think those disruptions will be less frequent industrial organization economists.
in the future. After more than 30 years, it seems
unlikely that airline losses are due entirely to a
series of unfortunate exogenous events relative
to what management and investors should have
expected. 5
All price calculations in this paper include average bag-
Throughout deregulation, the legacy carriers gage fees and cancellation fees by airline and so account for
the recent rise in revenues from these sources. 
have maintained much higher costs than LCCs, 6
There is a lengthy literature on the impact of airline alli-
but the price premia they have been able to charge ances that expand network effects. See Olivier Armantier
have declined by more than 60 percent over the and Oliver Richard (2008) and citations therein. 
VOL. 101 NO. 3 Why Can’t US Airlines Make Money?
REFERENCES
Armantier, Olivier, and Richard Olivier. 2008.
“Domestic Airline Alliances and Consumer
Welfare.” RAND Journal of Economics, 39(3),
875–904.
Borenstein, Severin. 1996. “Repeat-Buyer Pro-
grams in Network Industries.” In Networks,
Infrastructure, and the New Task for Regula-
tion, ed. Werner Sichel and Donald L. Alex-
ander, 137–62. Ann Arbor, MI: University of
Michigan Press.
Borenstein, Severin. 2011. “On the Persistent
Financial Losses of U.S. Airlines: A Pre-
liminary Exploration.” National Bureau of
237
­Economic Research Working Paper 16744.
Borenstein, Severin, and Nancy L. Rose. 2007.
“How Airline Markets Work . . . Or Do They?
Regulatory Reform in the Airline Industry.”
National Bureau of Economic Research Work-
ing Paper 13452.
Lederman, Mara. 2007. “Do Enhancements to
Loyalty Programs Affect Demand? The Impact
of International Frequent Flyer Partnerships on
Domestic Airline Demand.” RAND Journal of
Economics, 38(4): 1134–58.
Lederman, Mara. 2008. “Are Frequent-Flyer Pro-
grams a Cause of The “Hub Premium”?” Jour-
nal of Economics and Management Strategy,
17(1): 35–66.